1 - Garofalo-etal.pdf

MiR-34a/c-Dependent PDGFR-α/β Downregulation InhibitsTumorigenesis and Enhances TRAIL-Induced Apoptosisin Lung CancerMichela Garofalo1*☯, Young-Jun Jeon1☯, Gerard J. Nuovo1,2, Justin Middleton1, Paola Secchiero3, PoojaJoshi1, Hansjuerg Alder1, Natalya Nazaryan1, Gianpiero Di Leva1, Giulia Romano1, Melissa Crawford4,Patrick Nana-Sinkam4, Carlo M. Croce1* Department of Molecular Virology, Immunology and Medical Genetics, Comprehensive Cancer Center, the Ohio State University, Columbus, Ohio, UnitedStates of America, Phylogeny, Inc., Columbus, Ohio, United States of America, Department of Morphology and Embryology, Human Anatomy Section,University of Ferrara, Ferrara, Italy, Pulmonary, Allergy, Critical Care and Sleep Medicine, The Ohio State University Comprehensive Cancer Center,Columbus, Ohio, United States of America Abstract Lung cancer is the leading cause of cancer mortality in the world today. Although some advances in lung cancertherapy have been made, patient survival is still poor. MicroRNAs (miRNAs) can act as oncogenes or tumor-suppressor genes in human malignancy. The miRfamily consists of tumor-suppressive miRNAs, and its reducedexpression has been reported in various cancers, including non-small cell lung cancer (NSCLC). In this study, wefound that miR-34a and miR-34c target platelet-derived growth factor receptord growth factor receptor alpha and beta (PDGFR-α andPDGFR-β), cell surface tyrosine kinase receptors that induce proliferation, migration and invasion in cancer. MiR-34aand miR-34c were downregulated in lung tumors compared to normal tissues. Moreover, we identified an inversecorrelation between PDGFR-α/β and miR-34a/c expression in lung tumor samples. Finally, miR-34a/c overexpressionor downregulation of PDGFR-α/β by siRNAs, strongly augmented the response to TNF-related apoptosis inducingligand (TRAIL) while reducing migratory and invasive capacity of NSCLC cells. Citation: Garofalo M, Jeon Y , Nuovo GJ, Middleton J, Secchiero P, et al. (2013) MiR-34a/c-Dependent PDGFR-α/β Downregulation Inhibits Tumorigenesisand Enhances TRAIL-Induced Apoptosis in Lung Cancer. PLoS ONE 8(6): e67581. doi:10.1371/journal.pone.Editor: Noriko Gotoh, Institute of Medical Science, University of Tokyo, JapanReceived November 10, 2012; Accepted May 23, 2013; Published June 21, 2013Copyright: © Garofalo et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permitsunrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Funding: This work was supported by National Institutes of Health CA152758. The authors are grateful for research support from The Ohio StateUniversity Targeted Investment in Excellence Award. MG is a recipient of the Kimmel Scholar Award 2011. The funders had no role in study design, datacollection and analysis, decision to publish, or preparation of the manuscript. Competing interests: GJN is an employee of Phylogeny, Inc. There are no patents, products in development or marketed products to declare. This doesnot alter the authors' adherence to all the PLOS ONE policies on sharing data and materials. * E-mail: carlo.croce@osumc.edu (CC); michela.garofalo@osumc.edu (MG) ☯ These authors contributed equally to this work. Introduction Lung cancer is the most common cause of cancer deathworldwide [1]. Despite years of research, the prognosis forpatients with lung cancer remains dismal. The most frequenttype, non-small cell lung cancer (NSCLC) (85%), shows anoverall five year survival of 15%. Isoforms of platelet-derivedgrowth factor receptor (PDGFR) and its ligand, PDGF,constitute a family of receptors and ligands involved inproliferative and prosurvival signaling within cells. The PDGFR/PDGF system includes two receptors (PDGFR-α and PDGFR-β) and four ligands (PDGFA, PDGFB, PDGFC, and PDGFD).Ligand binding induces receptor dimerization, enablingautophosphorylation of specific tyrosine residues andsubsequent recruitment of a variety of signal transduction molecules. PDGFR regulates normal cellular growth anddifferentiation [2] and expression of activated PDGFR promotesoncogenic transformation [3]. Numerous in vitro and in vivostudies showed that inhibition of PDGFR-α signaling disruptscancer cell survival in the subset of gastrointestinal stromaltumors (GISTs) with activating PDGFR-α mutations [4,5]. In arecent study of NSCLC patient samples, activatedPDGFR-α was detected in 13% of cases [6], suggesting that asubset of these patients might might benefit from tbenefit from therapies directedagainst PDGFR-α. Moreover, PDGFR-α overexpression hasbeen observed in metastatic versus nonmetastaticmedulloblastoma patient samples and disrupting PDGFR-αfunction inhibited the metastatic potential of medulloblastomacells in vitro [7]. Given its progrowth role in cell signaling,PDGFR-α has become an attractive therapeutic target in a PLOS ONE | www.plosone.org June | Volume | Issue | e-J number of human malignancies. In non–small cell lung cancer,cytoplasmic PDGFR-α expression by tumor is a negativeprognostic indicator [8], confirming that the PDGF axis may bebiologically relevant. All members of the PDGF family displaypotent angiogenic activity in vivo, and from this point of view,PDGF-B/PDGFRβ axis was the most extensive evaluated. Inthe null mice it was shown that PDGF-B and PDGFRβ arecritically involved in vascular development. The role of PDGF/PDGFR in vascular development is supported by knockoutexperiments [9,10]. MicroRNAs (miRNAs), a class of ~ ntendogenous RNAs, are important regulators of geneexpression and have been implicated in the regulation ofcritical processes that are deregulated in cancer cells, asproliferation [11] differentiation [12] and apoptosis [13].Alterations in miRNA expression in cancer have beendocumented in numerous studies and suggest that miRNAscritically contribute to the cancer cell phenotype [14,15].Furthermore, some miRNA-encoding genes have beenclassified as oncogenic or tumor suppressive genes accordingto their function in cellular transformation and alteredexpression in tumors. In 2007, reports from several laboratories showed thatmembers of the miRfamily are direct ptargets, and thattheir upregulation induced apoptosis and cell-cycle arrest[16,17]. In mammalians, the miRfamily comprises threeprocessed miRNAs that are encoded by two different genes:miR-34a is encoded by its own transcript, whereas miR-34band miR-34c share a common primary transcript. Moreover,the promoter region of miR-34a, miR-34b and miR-34ccontains CpG islands and aberrant CpG methylation reducesmiRfamily expression in multiple types of cancer [18,19,20]. In this study, we show that miR-34a and miR-34c, arestrongly downregulated in NSCLC cells and lung tumorswhereas they are highly expressed in normal lung tissues.Moreover, miR-34a and miR-34c, by targeting PDGFR-α andPDGFR-β, increase TRAIL-induced apoptosis and decreaseinvasiveness of lung cancer cells. Furthermore, our resultssuggest, for the first time, that combinatory treatment of TRAILand PDGFR inhibitors could be effective for anti-NSCLCtherapy. Results MiR-34a and miR-34c target PDGFR-α and PDGFR-β 3’UTRs Among the miRNAs, miRfamily members play importanttumor suppressive roles, as they are directly regulated by p53and compose the pnetwork [16,17]. A previous studyindicated that miRmethylation was present in NSCLC andwas significantly related to an unfavorable clinical outcome[20]. First, we analyzed by Real Time PCR (qRT-PCR)miR-34a, -34b and -34c expression in different NSCLC celllines with pWT, mutant or null (Figure S1a in File S1).MiR-34a, -34b and -34c had low or absent expression in all thefive cell lines (Figure S1b in File S1). To identify miR-34a, -34b,and -34c targets, we performed a bioinformatics search(Targetscan, Pictar) for putative mRNA targets. Among thecandidate targets, 3’ UTR of human PDGFR-α and PDGFR-β contained regions (PDGFR-α nucleotides 2670–2676;2699-2705; PDGFR-β nucleotides 1535-1541) that matchedthe seed sequences of hsa-miR-34a, -34b and -34c (Figure1a). PDGFR-α and PDGFR-β have been reported to beoverexpressed and related to poor outcome in lung cancer [21].To verify whether PDGFR-α and PDGFR-β were direct targetsof miR-34a, -34b and -34c, PDGFR-α 3’ UTR, containing twomiR-34a, -34b and -34c binding sites and PDGFR-β 3’ UTR,containing one miR-34a, -34b and -34c binding site (Figure1a,b), were cloned downstream of the luciferase open readingframe. Interestingly, increased expression of miR-34a andmiR-34c, and not miR-34b, upon transfection, confirmed byqRT-PCR (data not shown), significantly decreased luciferaseactivity, indicating a direct interaction between the miRNAs andPDGFRα and PDGFRβ 3’ UTRs (Figure 1c,d). Target generepression was rescued by mutations in the complementaryseed sites (Figure 1b,c,d). Taken together the results indicatethat miR-34a and miR-34c and not miR-34b directly targetPDGFR-α and PDGFR-β. MiR-34a and miR-34c are inversely related to PDGFR-α/β expression in vitro and in vivo Next, we analyzed the consequences of the ectopicexpression of miR-34a and -34c in Caluand Hcells. Wechose these two cell lines because of the high expressionlevels of PDGFR-α (Hand Calu-6) and PDGFR-β (Calu-6)(data not shown). Increased expression of miR-34a andmiR-34c upon transfection was confirmed by qRT-PCR (datanot shown) and then the effects on PDGFR-α and PDGFR-βmRNA and protein levels were analyzed by qRT-PCR andwestern blot. MiR-34a and -34c (and not miR-34b)overexpression significantly reduced PDGFR-α and PDGFR-βmRNAs (Figure 2a) and the endogenous protein levels,compared to the cells transfected with a scrambled pre-miR(Figure 2b). The expression levels of the four PDGF ligands(PDGFA, PDGFB, PDGFC, PDGFD) after miR-34a andmiR-34c enforced expression were also evaluated in bothCaluand Hcells. PDGFD was barely expressed in bothcell lines; we did not find any variation of the expression ofPDGFA, PDGFB and PDGFC (data not shown). In summary,these results supported the bioinformatics predictionsindicating PDGFR-α and PDGFR-β 3’ UTRs as targets ofmiR-34a and -34c. To verify the downregulation of miR-34a and -34c also invivo, lung tumors (among adenocarcinoma and squamouscell carcinoma) and the adjacent histologically normal lungtissues were analyzed for miR-34a and -34c expression. Asshown in Figure 3a and b, miR-34a and miR-34c expressionwas lower in the tumor compared to the normal samples(Figure 3a,b). Moreover, we analyzed miR-34a, 34c and PDGFR-α/βmRNA expression in primary human lung tumor specimensin comparison with normal tissues. MiR-34a and -34c werealmost undetectable in the tumor and highly expressed in thenormal lung samples tested (Figure 3c). Remarkably, aninverse correlation between miR-34a/c and PDGFR-α andPDGFR-β was observed (Figure 3d). To further corroboratethese findings, in situ hybridization (ISH) analysis was miR-34a/c Induce TRAIL Sensitivity PLOS ONE | www.plosone.org June | Volume | Issue | eperformed using 5’-dig-labeled LNA probes on lung tumors andnormal tissues, followed by immunohistochemistry (IHC) forPDGFR-α and PDGFRβ. Most lung cancer cells showed lowexpression of miR-34a and high expression of PDGFR-α/β,whereas the adjacent non-malignant lung expressed PDGFR-αrarely and abundantly expressed miR-34a. MiR-34a andPDGFR-α/β expression in the majority of the differenttumors analyzed was basically mutually exclusive (Figure 4a,band Figure Sin File S1). MiR-34a and miR-34c overcome TRAIL resistance ofNSCLC cells through PDGFR-α and PDGFR-βdownregulation TNF-related apoptosis-inducing ligand (TRAIL) is a memberof the tumor necrosis factor family, known to induce apoptosisin a variety of different tumor types. TRAIL is able tospecifically induce cell death in cancer cells while sparingnormal cells and is currently being tested as a promising anti-tumor agent in clinical trials [22]. However, many tumorsincluding NSCLC are resistant to TRAIL thus limiting itstherapeutic application. Since PDGFR-α and PDGFR-βregulate the PI3K/Akt and ERK1/pathways [23,24,25], we Figure 1. MiR-34a and miR-34c target PDGFR-α and PDGFR-β 3’ UTRs. (a) PDGFR-α and PDGFR-β 3’ UTRs contain,respectively, two and one predicted miR-34a, -34b and -34c binding sites. In the figure the alignment of the seed regions ofmiR-34a/c with PDGFR-α and PDGFR-β 3’ UTRs is shown. The sites of target mutagenesis are indicated in red. _deletednucleotides. (b) PGLcontrol-PDGFR-α constructs containing two PDGFR-α binding sites (in red). Deletion of one of the twoPDGFR-α sites was used to generate the mutant luciferase palsmids. (c–d) PDGFR-α and PDGFR-β 3’ UTRs are direct targets ofmiR-34a and miR-34c. pGL3-PDGFR-α and pGL3-PDGFR-β luciferase constructs, containing a wild-type (left side of thehistograms) or mutated (right side of the histograms) PDGFR-α and PDGFR-β 3’ UTRs, were transfected into Calucells. Relativerepression of firefly luciferase expression was standardized to a transfection control. The reporter assays were performed threetimes with essentially identical results. *P<0.0001, **P<by two tailed Student’s t test.doi: 10.1371/journal.pone.0067581.gmiR-34a/c Induce TRAIL Sensitivity PLOS ONE | www.plosone.org June | Volume | Issue | enext examined, by immunostaining, the expression and/oractivation of some of the proteins involved in these pathwaysfollowing miR-34a and miR-34c enforced expression orPDGFR-α/β silencing by siRNAs. As shown in Figure 5a,phosphorylation levels of ERKs decreased after miR-34a andmiR-34c enforced expression compared to cells transfectedwith the control miR. PDGFR-α silencing reduced the activationof both Akt and ERK1/(Figure 5b). We previouslydemonstrated that the PI3K/AKT pathway plays a key role inTRAIL-induced apoptosis [26], therefore the effects of miR-34aand miR-34c overexpression on cell survival and TRAILresistance of NSCLC were examined. First, we performed aproliferation assay on Caluand HTRAIL semi-resistantcells after enforced expression of miR-34a and miR-34c. 48hafter transfection cells were exposed to TRAIL for 24h and thencell proliferation was assessed using an MTT assay. Calu-6and Hcells overexpressing miR-34a and -34c showed a significant lower proliferation rate compared to the control cells(Figure S3a in File S1). Moreover, caspase 3/assay revealedan increase in TRAIL sensitivity after miR-34a and -34cenforced expression or PDGFR-α/β silencing, compared to thecells transfected with a scrambled miR or siRNA (Figure 5c,d). Furthermore, overexpression of PDGFR-α and PDGFR-β inHTRAIL-sensitive cells, increased the resistance to thedrug (Figure 5e and Figure S3c in File S1). Conversely,treatment of Calucells with a PDGFR inhibitor significantlyincreased the sensitivity to TRAIL-induced apoptosis (Figure 5fand Figure S3d in File S1). Intriguingly, overexpression ofPDGFR-α or PDGFR-β (using two plasmids containing only thecoding sequences and not the 3’ UTRs of PDGFR-α/β) alongwith miR-34a or miR-34c, decreased the sensitivity to TRAIL-induced apoptosis, as assessed by both MTT and caspase 3/7assay (Figure Sin File S1). The results suggest that PDGFR-α and PDGFR-β play an important role in TRAIL-induced Figure 2. MiR-34a and miR-34c reduce PDGFR-α and PDGFR-β mRNA and protein levels. (a) qRT-PCR showing PDGFR-αand PDGFR-β mRNAs downregulation in Caluand Hcells after miRand miR-34c but not miR-34b enforced expression(b) miR-34a and miR-34c enforced expression decreases endogenous levels of PDGFR-α/β protein levels in Hand Calu-6NSCLC. Cells were transfected with either scrambled, miR-34a or miR-34c for 72h. PDGFR-α and PDGFR-β expression wasassessed by western blot. Loading control was obtained using GAPDH antibody. *P<0.05, **P<by two tailed Student’s t test.doi: 10.1371/journal.pone.0067581.gmiR-34a/c Induce TRAIL Sensitivity PLOS ONE | www.plosone.org June | Volume | Issue | eFigure 3. MiR-34a/c and PDGFR-α/β are inversely correlated in normal and tumor lung tissue samples. (a–b) qRT-PCR onlung tumor and normal tissues. MiR-34a and miR-34c are downregulated in the tumors compared to the normal lung tissues. (c)Box plots showing miR-34a and miR-34c expression in lung normal and cancer tissues. (d) XY scatter plots showing inversecorrelation between miR-34a/c and PDGFR-α/β. Two-tailed Student’s t test was used to verify the significance. P<0.05.doi: 10.1371/journal.pone.0067581.gmiR-34a/c Induce TRAIL Sensitivity PLOS ONE | www.plosone.org June | Volume | Issue | eapoptosis and that PDGFR inhibitor can sensitize NSCLC cellsto TRAIL with important therapeutic consequences. PDGFR-α/β downregulation by miR-34a and miR-34cinhibits migration and invasiveness of NSCLC cells Because PDGFR-α/β regulate the PI3K/AKT pathway,notably involved in migration and invasion of different tumors[27,28], we investigated if miR-34a/c could influence NSCLCmigration and invasion through PDGFR-α and PDGFR-βdownregulation. To directly test the functional role of miR-34a/cin tumorigenesis, we overexpressed these two miRNAs or silenced PDGFR-α/β in Caluor Hcells. Intriguingly, weobserved a significant decrease of the migratory and invasivecapabilities of Caluand Hcells after miR-34a ormiR-34c overexpression (Figure 6a) as well after PDGFR-αand PDGFR-β downregulation (Figure 6b), confirmed also byscratch-wound assay (Figure 6c). To further verify thatPDGFR-α and PDGFR–β were involved in tumorigenesis ofNSCLC cells, miR-34a and -34c were transfected in Calu-6cells alone or in combination with a plasmid overexpressingonly the coding sequence and not the 3’ UTR of PDGFR-α andPDGFR-β. MiR-34a/c enforced expression reduced migrationand invasion of Calucells but overexpression of PDGFR-α or Figure 4. MiR-34a/c and PDGFR-α/β co-expression in vivo. (a–b) Immunohistochemistry and in situ hybridization on lungcancer tissues samples. MiR-34a (blue) and PDGFR-α/β (brown/red, respectively in RGB and each fluorescent red in Nuanceconverted image) expression was inversely related in lung cancers and the adjacent normal lung tissues. These serial sections wereanalyzed for miR-34a expression by in situ hybridization, followed by immunohistochemistry for PDGFR-α/β. (a) Representativeexample: Co-expression analysis of miR-34a and PDGFR-α. Note lack of expression in the merged image (panel b) (fluorescentyellow = co-expression). (b) Representative example: miR-34a= blue (panel a), PDGFR-β=red (panel b), co-expression= yellow(panel c). RGB= Regular Green Blue image of the ISH/lHC reaction shown in panels a-c. Scale bar indicates μm. Themagnification is the same for all the panels.doi: 10.1371/journal.pone.0067581.gmiR-34a/c Induce TRAIL Sensitivity PLOS ONE | www.plosone.org June | Volume | Issue | eFigure 5. MiR-34a and miR-34c overexpression or PDGFR-α/β silencing increases the response of NSCLC cells to TRAIL-induced apoptosis. (a) Western blot in Calucells after miR-34a, -34b and -34c forced expression. MiR-34a or miR-34c and notmiR-34b forced expression decreases PDGFRβ expression levels and reduces the activation of the ERK1/2. (b) Western blotshowing the inactivation of the Akt and ERKs pathways after PDGFR-α silencing. (c) Caspase 3/assay. MiR-34a and -34cenforced expression in Caluand Hsemi-resistant cells, increases the response to TRAIL-induced apoptosis. (d) Caspase3/assay showing that PDGFR-α or PDGFR-β silencing increases the response to TRAIL-induced apoptosis. (e) PDGFR-α orPDGFR-β overexpression in HTRAIL-sensitive cells decreases the response to the drug. (f) Combined treatment of PDGFRinhibitor (μM) and different TRAIL concentrations (0-100ng/ml) for 24h sensitizes NSCLC cells to TRAIL-induced apoptosis.*P<0.001, ** P<0.05.doi: 10.1371/journal.pone.0067581.gmiR-34a/c Induce TRAIL Sensitivity PLOS ONE | www.plosone.org June | Volume | Issue | ePDGFR-β, along with the two microRNAs, partially restored themigration and invasion capabilities, suggesting that miR-34a/cregulate NSCLC tumorigenesis, at least in part, throughPDGFR-α/ β (Figure d,e). Discussion Lung cancer is the leading cause of cancer death in bothmen and women worldwide1. The American cancer Societyestimates 156,deaths from lung cancer in the UnitedStates for alone [29]. Non-small cell lung cancer (NSCLC)accounts for the majority of all lung cancer cases and is aleading cause of cancer mortality [30]. The high mortality rate associated with lung cancer hasprompted numerous exhaustive efforts to identify noveltherapeutic targets and treatment modalities for this deadlydisease. Platelet-derived growth factor receptors (PDGFRs)and their ligands, platelet-derived growth factors (PDGFs) playcritical roles in mesenchymal cell migration and proliferation.Abnormalities of PDGFR/PDGF are thought to contribute to anumber of human diseases and especially malignancy [31]. MicroRNAs are small noncoding RNAs that showderegulation in most cancers. There is growing evidence thatthey play substantial roles in the pathogenesis and prognosisof human malignancies and in the resistance tochemotherapeutic drugs [32,33]. Tumor necrosis factor-relatedapoptosis-inducing ligand (TRAIL) triggers apoptosis in tumor Figure 6. MiR-34a and miR-34c overexpression or PDGFR-α/β silencing decreases migratory and invasive capacity ofNSCLC cells. (a) MiR-34a and -34c enforced expression reduces migratory and invasive capabilities of Hcells. (b) PDGFR-αand PDGFR-β silencing reduces migratory and invasive capabilities of Calucells. RFU= Relative Fluorescence Units. (c)Representative photographs of scratched areas of the confluent monolayer of Calucells transfected with miR-34a/c or controlmiRNA (miR-Ctr) at 0h, 12h and 24h after wounding with a pipet tip. Scale bar, μm. The magnification is the same for all thepanels. (d–e) PDGFR-α and PDGFR-β overexpression partially rescues migratory and invasive capabilities of Calucells. *P<0.05.doi: 10.1371/journal.pone.0067581.gmiR-34a/c Induce TRAIL Sensitivity PLOS ONE | www.plosone.org June | Volume | Issue | ecells, but when used alone, it is ineffective at treating TRAIL-resistant tumors. This resistance is challenging for TRAIL-based anti-cancer therapies. In this study, we found thatmiR-34a and miR-34c are strongly downmodulated in bothNSCLC cells and lung tumors compared to normal tissues.Enforced expression of miR-34a and miR-34c downregulatedPDGFR-α and PDGFR-β mRNA and protein levels. Luciferaseand western blot experiments demonstrated that PDGFR-α andPDGFR-β are direct targets of miR-34a and miR-34c but not ofmiR-34b. The resistance of many types of cancer toconventional chemotherapies is a major factor underminingsuccessful cancer treatment. AKT activation also contributes totumorigenesis and tumor metastasis, and as shown mostrecently, resistance to chemotherapy [26,34]. As a result, bothin vitro and in vivo studies combining small molecule inhibitorsof the PI3K/Akt pathway with standard chemotherapy haveproven successful in attenuating chemotherapeutic resistance.Specifically, inhibiting AKT activity may be a valid approach totreat cancer and increase the efficacy of chemotherapy. Protein kinases are major regulators of most cellularsignaling pathways. Among them, receptor tyrosine kinases(RTKs), such as PDGFR, play pivotal roles in promotingcellular growth and proliferation by transducing extracellularstimuli to intracellular signaling circuits [35]. A prominentcomponent of the intracellular signaling machinery is the PI3K/Akt(PKB)/mammalian target of rapamycin (PI3K/Akt[PKB]/mTOR) pathway [36,37]. Aberrant activation of this pathway bymutation of any of multiple genes is known to occur in themajority of human cancers through various mechanisms[38,39]. In a previous work [26], we demonstrated that MET,through the activation of the PI3K/AKT pathway, inducedtumorigenesis and TRAIL resistance in NSCLC. Therefore, wehypothesized that PDGFR-α/β, through the activation of theAKT pathway should be involved in TRAIL-induced apoptosis.Indeed, overexpression of miR-34a and miR-34c ordownregulation of PDGFR-α/β by siRNAs, highly increased theresponse of semi-resistant NSCLC cells to TRAIL-inducedapoptosis. Importantly, combined treatment of a PDGFRinhibitor with TRAIL, increased apoptosis and reduced cellproliferation, as assessed by caspase 3/assay and MTTassays. Taken together, the results suggest that combinedtreatment of TRAIL with PDGFR inhibitors could sensitize asubset of lung tumors, expressing the PDGF receptors, to thedrug. Moreover, it is well known that the PI3K/AKT, as well theERK1/pathways regulate cellular migration and invasion ofdifferent cancers [40,41]. Here, we reported that miR-34a andmiR-34c overexpression or PDGFR-α/β silencing inhibited themigration and invasion capacity of Caluand Hcells,compared to cells transfected with a scrambled miR or siRNAcontrol. Enforced expression of PDGFR-α or PDGFR-β partiallyrestored NSCLC migration and invasion supporting that theregulation of the expression of these receptors by miR-34a/cplays an important role in NSCLC tumorigenesis. However, werecognize that other miR-34a/c targets including c-Met [16] andAXL [42] could also be involved. While this manuscript was inpreparation Silber et al. reported that miR-34a expression waslower in proneural gliomas compared to other tumor subtypesand identified PDGFR-α as a direct target of miR-34a [43]. Here, we report that not only miR-34a but miR-34c alsodownregulates PDGFR-α in NSCLC cells. Moreover, wedemonstrate that PDGFR-β is a miR-34a/c direct target whilewe did not see any significant effect on the expression ofPDGFR-α and PDGFR-β after miR-34b enforced expression.Remarkably, our study shows that inhibition or downregulationof PDGFR-α and PDGFR-β by miR-34a/c antagonizestumorigenicity and increases sensitivity to TRAIL-induced celldeath with important therapeutic application for future anti-tumor therapy of lung cancer. Materials and Methods Lung cancer cell lines and tissue samplesHuman H460, A549, H1299, Hcell lines were grown in RPMI medium containing 10% heat-inactivated fetal bovineserum (FBS) and with 2mM L-glutamine and 100Uml-1penicillin-streptomycin. Calucells were grown in MEMsupplemented with 10% fetal bovine serum, 2mM L-glutamineand 100Umlpenicillin–streptomycin. lung tumors (includingadenocarcinoma and squamous cell carcinoma) and theirnormal counterparts were kindly provided by Dr. S.P. Nana-Sinkam, Pulmonary, Allergy, Critical Care and Sleep Medicine,The Ohio State University Comprehensive Cancer Center,Columbus, OH. lung normal and tumor tissue samples wereprovided from the Department of Pathology, Ohio StateUniversity. All human tissues were obtained according to aprotocol approved by the Ohio State Institutional ReviewBoard. Luciferase assayThe 3’ UTRs of the human PDGFRA and PDGFRB genes, were PCR amplified using the following primers: PDGFR-α FW 5’ TCTAGACCGGCCTGAGAAACACTATTTGTG 3’ PDGFR-α RW 5’TCTAGAACATGAACAGGGGCATTCGTAATACA 3″ PDGFR-β FW 5’TCTAGAAAAGAGGGCAAATGAGATCACCTCCTGCA 3’ PDGFR-β RW 5’TCTAGATATTGAGAACCCACTCTCCCTCCTTGGA 3’ and cloned downstream of the Renilla luciferase stop codonin pGLcontrol vector (Promega). These constructs were usedto generate, by inverse PCR, the p3’-UTRs- mutant-plasmidsusing the following primers: PDGFR-α mutFW 5’ACTGCCAAAACATTTATGACAAGCTGTATCGCCTCG 3’ PDGFR-α mutRW 5’CGAGGCGATACAGCTTGTCATAAATGTTTTGGCAGT PDGFR-αmutFW: 5’ACTGCCAAAACATTTATGACAAGCTGTATGGTCGTTTATATTT 3’ PDGFR-αmutRW:5’AAATATAAACGACCATACAGCTTGTCATAAATGTTTTGGCAGT 3’ miR-34a/c Induce TRAIL Sensitivity PLOS ONE | www.plosone.org June | Volume | Issue | ePDGFR-β Mut FW 5'-ATGGGGGTATGGTTTTGTCAGACCTAGCAGTGAC-3' PDGFR-β Mut RW 5'-GTCACTGCTAGGTCTGACAAAACCATACCCCCAT-3' Calucells were cotransfected with 1μg of p3’UTR-PDGFR-α, p3’UTR-PDGFR-β or with p3’UTRmut-PDGFR-α andp3’UTRmut-PDGFR-β, μg of a Renilla luciferase expressionconstruct pRL-TK (Promega) by using Lipofectamine 2000(Invitrogen). Cells were harvested 24h post-transfection andassayed with Dual Luciferase Assay (Promega) according tothe manufacturer’s instructions. Three independentexperiments were performed in triplicate. Western Blot AnalysisTotal proteins from NSCLC were extracted with radioimmunoprecipitation assay (RIPA) buffer (0.15mM NaCl,0.05mM Tris-HCl, pH 7.5, 1% Triton, 0.1% SDS, 0.1% sodiumdeoxycholate and 1% Nonidet P40). Sample extract (μg)was resolved on 7.5–12% SDS–polyacrylamide gels (PAGE)using a mini-gel apparatus (Bio-Rad Laboratories) andtransferred to Hybond-C extra nitrocellulose. Membranes wereblocked for 1h with 5% nonfat dry milk in Tris-buffered salinecontaining 0.05% Tween 20, incubated overnight with primaryantibody, washed and incubated with secondary antibody, andvisualized by chemiluminescence. The following primaryantibodies were used: anti-PDGFR-α, anti-PDGFR-β, anti-ERK1/2, anti-p-ERKs, anti-pAKT, anti-total AKT, anti-GAPDHantibodies (Cell Signaling). A secondary anti-rabbit or anti-mouse immunoglobulin G (IgG) antibody peroxidase conjugate(Chemicon) was used. Real-time PCRReal-time PCR was performed using a standard TaqMan PCR Kit protocol on an Applied Biosystems 7900HT SequenceDetection System (Applied Biosystems). The μl PCRreaction included μl RT product, μl TaqMan UniversalPCR Master Mix (Applied Biosystems), mM TaqMan probe,mM forward primer and mM reverse primer. Thereactions were incubated in a 96-well plate at °C for min,followed by cycles of °C for s and °C for min. Allreactions were run in triplicate. The threshold cycle (CT) isdefined as the fractional cycle number at which thefluorescence passes the fixed threshold. The comparative CTmethod for relative quantization of gene expression (AppliedBiosystems) ) was used to determine miRNA expression levels.The y axis represents the 2(-ΔCT), or the relative expression ofthe different miRs. MiRs expression was calculated relative toUand UrRNA. Experiments were carried out in triplicatefor each data point, and data analysis was performed by usingsoftware (Bio- Rad). Cell death and cell proliferation quantificationFor detection of caspase 3/activity, cells were cultured in 96-well plates, in triplicate, treated with TRAIL (100-150ng/ml)and analyzed using Caspase-Glo 3/Assay kit (Promega)according to the manufacturer’s instructions. Continuousvariables are expressed as mean values ± standard deviation (s.d.). Cell viability was examined with 3- (4,5-dimethylthiazol-2-yl)-2,5-dipheniltetrazolium bromide (MTT)-Cell Titer AQueous One Solution Cell Proliferation Assay(Promega), according to the manufacturer’s protocol.Metabolically active cells were detected by adding μl of MTTto each well. After h of incubation, the plates were analyzedin a Multilabel Counter (Bio-Rad Laboratories). Anti-PDGFR-α and anti-PDGFR-β siRNAs transfectionCells were cultured to 50% confluence and transiently transfected for 72h using Lipofectamine with nM anti-PDGFR-α and/or with 100nM anti-PDGFR-β SMARTpoolsiRNAs or control siRNAs (Dharmacon), a pool of four targetspecific 20–nt siRNAs designed to knock down geneexpression. MiRNA locked nucleic acid in situ hybridization offormalin fixed, paraffin-embedded tissue section In situ hybridization (ISH) was carried out on deparaffinizedhuman lung tissues using previously published protocol[44],which includes a digestion in pepsin (mg/ml) for 30minutes. The sequence of the probe containing the dispersedlocked nucleic acid (LNA) modified bases with digoxigeninconjugated to the 5’ end was: 5’ACAACCAGCTAAGACACTGCCA 3’. The probe cocktail andtissue miRNA were co-denatured at °C for minutes,followed by hybridization at °C overnight and a stringencywash in 0.2X SSC and 2% bovine serum albumin at °C for 10minutes. The probe-target complex was seen due to the actionof alkaline phosphatase on the chromogen nitrobluetetrazolium and bromochloroindolyl phosphate (NBT/BCIP).Negative controls included the use of a probe, which shouldyield a negative result in such tissues (scrambled miRNA). Nocounterstain was used, to facilitate co-labeling for PDGFR-αand PDGFR-β protein. After in situ hybridization for themiRNAs, as previously described as previously described as previously described (Nuovo et al., 2009), theslides were analyzed for immunohistochemistry using theoptimal conditions for PDGFR-α (1:100, cell conditioning for 30minutes) and PDGFR-β (1:200, cell conditioning for 30minutes). For the immunohistochemistry, we used theUltrasensitive Universal Fast Red or DAB systems fromVentana Medical Systems. The percentage of tumor cellsexpressing PDGFR-α, PDGFR-β and miR-34a, was thenanalyzed with emphasis on co-localization of the respectivetargets. Co-expression analysis was done with the Nuancesystem (Cambridge Research Institute) per the manufacturer’srecommendations. Bioinformatics analysisBioinformatics analysis was performed by using these specific programs:Targetscan1, Pictar2, RNhybrid 3: http://www.targetscan.org/http://pictar.bio.nyu.edu/http://bibiserv.techfak.uni-bielefeld.de/ miR-34a/c Induce TRAIL Sensitivity PLOS ONE | www.plosone.org June | Volume | Issue | ePDGFR-α and PDGFR-β plasmidscDNA-PDGFR-α (Cat. Number MHS1010-9205933) and cDNA-PDGFR-β (Cat. MHS1010-7430189) were purchasedfrom Open Biosystems. HTRAIL-sensitive cells weretransfected with 1μg of each plasmid and proliferation andcaspase 3/assays were performed as previously described. Migration assayBriefly, Calucells were transfected with pcDNA-PDGFR-α, pcDNA-PDGFR-β and/or hsa-miR-34a and miR-34c,respectively. 24h after transfection, 2xcells in MEM mediasupplemented with 1% FBS were plated into the upperchambers of the Migration assay and RPMI supplemented with10% FBS were added into lower chambers to use as achemoattractant. After 24h, the upper chambers weretransferred into a new plate with detaching solutionscontainingCalcein AM for hour to measure the amount of the migratedcells. The fluorescence was analyzed at an excitationwavelength of nm and an emission wavelength of 520nm. Invasion assayBriefly, 2x Calucells transfected with pcDNA-PDGFR- α, pcDNA-PDGFR-β and/or miR-34a and miR-34c in RPMIsupplemented with 1% FBS were plated into upper chambersof Invasion assay with a 8-um pore size-polycarbonatemembrane. μl of MEM supplemented with 10% FBS wereadded into the lower chambers as a chemoattractant. After36-48h, the upper chambers were transferred into a new plateand were incubated with detaching solutions contained CalceinAM for hour to measure the amount of the invaded cells. Thefluorescence was analyzed at an excitation wavelength of 485nm and an emission wavelength of 520nm. Statistical analysisStudent’s t test was used to determine significance. All error bars represent the standard error of the mean. Statisticalsignificance for all the tests, assessed by calculating P-value,was < 0.05. Supporting Information File S1. Figure S1, NSCLC cell lines analyzed and p53status. (a) A panel of NSCLC cells with their pstatus is reported. (b) qRT-PCR showing low expression ofmiR-34a,-34b,-34c in different NSCLC cells. Figure S2, Co-expression analysis of miR-34a and PDGFR-α and PDGFR-β in lung tumor samples. Tables reporting the percentage ofmiR-34a, PDGFR-α and PDGFR-β expression observed in the(PDGFR-α) and (PDGFR-β) tumor samples analyzed(A case with 10% of the tumor cells + was scored as +). FigureS3, Enforced expression of miR-34a and miR-34c or PDGFR-α/β silencing increases the response to TRAIL-inducedapoptosis and reduces tumorigenicity of NSCLC cell. (a)Proliferation assay showing that miR-34a and -34c enforcedexpression in Caluand Hcells increases the responseto TRAIL-induced apoptosis. (b) MTT assay showing thatPDGFR-α or PDGFR-β silencing increases the response toTRAIL-induced apoptosis. (c) PDGFR-α or PDGFR-βoverexpression in HTRAIL-sensitive cells decreases theresponse to the drug as assessed by caspase 3/activity. (d)Combined treatment of PDGFR inhibitor (μM) and TRAIL for24h sensitizes NSCLC cells to TRAIL-induced apoptosis. * P<0.05. Figure S4, PDGFR-α or PDGFR-β overexpressionreduces the response to TRAIL-induced apoptosis. (a)Proliferation assay showing that miR-34a/c increase theresponse to TRAIL-induced apoptosis. Co-transfection ofmiR-34a/c with PDGFR-α/β significantly decreases theresponse to the drug. (b) PDGFR-α/β enforced expressionalong with miR-34a/c reduce the response to TRAIL-inducedapoptosis as assessed by caspase 3/assay. * P< 0.05.(PDF) Acknowledgements We thank Arianna Bottoni and P. Fadda for qRT-PCRassistance. Author Contributions Conceived and designed the experiments: MG. Performed theexperiments: MG YJJ GJN JM PJ NN GDL GR HA. Analyzedthe data: MG YJJ JM PJ HA. Contributed reagents/materials/analysis tools: MG CMC. Wrote the manuscript: MG. Providedthe TRAIL for the in vitro experiments: PS. Provided the RNAof normal and tumor lung samples: MC PNS. References 1. Gompelmann D, Eberhardt R, Herth FJ (2011) Advanced malignantlung disease: what the specialist can offer. Respiration 2: 111-23.PubMed: 21778793. 2. Ross R, Raines EW, Bowen-Pope DF (1986) The biology of platelet-derived growth factor. Cell 46(2): 155-69. doi:10.1016/0092-8674(86)90733-6. PubMed: 3013421. 3. Gazit A, Igarashi H, Chiu IM, Srinivasan A, Yaniv A et al. (1984)Expression of the normal human sis/PDGFcoding sequence inducescellular transformation. Cell 39(1): 89-97. doi:10.1016/0092-8674(84)90194-6. PubMed: 6091919. 4. Heinrich MC, Corless CL, Demetri GD, Blanke CD, von Mehren M et al.(2003) Kinase mutations and imatinib response in patients withmetastatic gastrointestinal stromal tumor. J Clin Oncol 21(23): 4342-9.doi:10.1200/JCO.2003.04.190. PubMed: 14645423. 5. Corless CL, Schroeder A, Griffith D, Town A, McGreevey L et al. (2005)PDGFRA mutations in gastrointestinal stromal tumorsmutations in gastrointestinal stromal tumors: frequency,spectrum and in vitro sensitivity to imatinib. J Clin Oncol 23(23):5357-64. doi:10.1200/JCO.2005.14.068. PubMed: 15928335. 6. Rikova K, Guo A, Zeng Q, Possemato A, Yu J et al. (2007) Globalsurvey of phosphotyrosine signaling identifies oncogenic kinases inlung cancer. Cell 131(6): 1190-203. doi:10.1016/j.cell.2007.11.025.PubMed: 18083107. 7. MacDonald TJ, Brown KM, LaFleur B, Peterson K, Lawlor C et al.(2001) Expression profiling of medulloblastoma: PDGFRA and theRAS/MAPK pathway as therapeutic targets for metastatic disease. NatGenet 29(2): 143-52. doi:10.1038/ng731. PubMed: 11544480. 8. Donnem T, Al-Saad S, Al-Shibli K, Andersen S, Busund LT et al. (2008)Prognostic impact of platelet-derived growth factors in non-small cell miR-34a/c Induce TRAIL Sensitivity PLOS ONE | www.plosone.org June | Volume | Issue | elung cancer tumor and stromal cells. J Thorac Oncol 3(9): 963-70. doi:10.1097/JTO.0b013e3181834f52. PubMed: 18758297. 9. Levéen P, Pekny M, Gebre-Medhin S, Swolin B, Larsson E et al. (1994)Mice deficient for PDGF B show renal, cardiovascular, andhematological abnormalities. Genes Dev 8: 1875–1887. doi:10.1101/gad.8.16.1875. PubMed: 7958863. 10. Soriano P (1994) Abnormal kidney development and hematologicaldisorders in PDGF beta-receptor mutant mice. Genes Dev 8: 1888–1896. doi:10.1101/gad.8.16.1888. PubMed: 7958864. 11. Wu N, Liu X, Xu X, Fan X, Liu M et al. (2011) MiR-373, a novelregulator of PPP6C, functions as an oncogene in hepatocellularcarcinoma. FEBS J 278: 2044–2054. doi:10.1111/j.1742-4658.2011.08120.x. PubMed: 21481188. 12. Annibali D, Gioia U, Savino M, Laneve P, Caffarelli E et al. (2012) Anew module in neural differentiation control: two microRNAsupregulated by retinoic acid, miRand -103, target the differentiationinhibitor ID2. PLOS ONE 7(7): e40269. doi:10.1371/journal.pone.0040269. PubMed: 22848373. 13. Chan JA, Krichevsky AM, Kosik KS (2005) MicroRNAis anantiapoptotic factor in human glioblastoma cells. Cancer Res 65: 6029–6033. doi:10.1158/0008-5472.CAN-05-0137. PubMed: 16024602. 14. Nishimura J, Handa R, Yamamoto H, Tanaka F, Shibata K et al. (2012)microRNA-181a is associated with poor prognosis of colorectal cancer.Oncol Rep 28(6): 2221-6. PubMed: 23023298. 15. Volinia S, Calin GA, Liu CG, Ambs S, Cimmino A et al. (2006) AmicroRNA expression signature of human solid tumors defines cancergene targets. Proc Natl Acad Sci U S A 103(7): 2257-61. doi:10.1073/pnas.0510565103. PubMed: 16461460. 16. He L, He X, Lim LP, de Stanchina E, Xuan Z et al. (2007) A microRNAcomponent of the ptumour suppressor network. Nature 447(7148):1130-4. doi:10.1038/nature05939. PubMed: 17554337. 17. Bommer GT, Gerin I, Feng Y, Kaczorowski AJ, Kuick R et al. (2007)p53-mediated activation of miRNAcandidate tumor-suppressorgenes. Curr Biol 17(15): 1298-307. doi:10.1016/j.cub.2007.06.068.PubMed: 17656095. 18. Lodygin D, Tarasov V, Epanchintsev A, Berking C, Knyazeva T et al.(2008) Inactivation of miR-34a by aberrant CpG methylation in multipletypes of cancer. Cell Cycle 7(16): 2591-600. doi:10.4161/cc.7.16.6533.PubMed: 18719384. 19. Chim CS, Wan TS, Wong KY, Fung TK, Drexler HG et al. (2011)Methylation of miR-34a, miR-34b/c, miR-124and miRin Ph-negative myeloproliferative neoplasms. J Transl Med 9: 197. doi:10.1186/1479-5876-9-197. PubMed: 22082000. 20. Wang Z, Chen Z, Gao Y (2011) DNA hypermethylation ofmicroRNA-34b/c has prognostic value for stage I non-small cell lungcancer. Cancer Biol Ther 11(5): 490-6. doi:10.4161/cbt.11.5.14550.PubMed: 21383543. 21. Rossi G, Cavazza A (2005) Role of chemotherapy and the receptortyrosine kinases KIT, PDGFRalpha, PDGFRbeta, and Met in large-cellneuroendocrine carcinoma of the lung. J Clin Oncol 23(34): 8774-85. 22. Mahalingam D, Oldenhuis CN, Szegezdi E, Giles FJ, de Vries EG et al.(2011) Targeting TRAIL towards the clinic. Curr Drug Targets 12(14):2079-2090. doi:10.2174/138945011798829357. PubMed: 21777191. 23. Feng H, Liu KW, Guo P, Zhang P, Cheng T et al. (2012) Dynamin 2mediates PDGFRα-SHP-2-promoted glioblastoma growth and invasion.Oncogene 31(21): 2691-2702. doi:10.1038/onc.2011.436. PubMed:21996738. 24. Dierov J, Xu Q, Dierova R, Carroll M (2002) TEL/platelet-derivedgrowth factor receptor beta activates phosphatidylinositol 3(PI3) kinaseand requires PIkinase to regulate the cell cycle. Blood 99(5):1758-65. doi:10.1182/blood.V99.5.1758. PubMed: 11861293. 25. Yu J, Deuel TF, Kim HR (2000) Platelet-derived growth factor (PDGF)receptor-alpha activates c-Jun NH2-terminal kinaseand antagonizesPDGF receptor-beta -induced phenotypic transformation. J Biol Chem275(25): 19076-82. doi:10.1074/jbc.M910329199. PubMed: 10777515. 26. Garofalo M, Di Leva G, Romano G, Nuovo G, Suh SS et al. (2009)miR-221&regulate TRAIL resistance and enhance tumorigenicitythrough PTEN and TIMPdownregulation. Cancer Cell 16(6): 498-509.doi:10.1016/j.ccr.2009.10.014. PubMed: 19962668. 27. Wu K, Fan J, Zhang L, Ning Z, Zeng J et al. (2012) PI3K/Akt to GSK3β/β-catenin signaling cascade coordinates cell colonization for bladder cancer bone metastasis through regulating ZEBtranscription. CellSignal 24(12): 2273-2282. doi:10.1016/j.cellsig.2012.08.004. PubMed:22906492. 28. Lehman HL, Van Laere SJ, van Golen CM, Vermeulen PB, Dirix LY etal. (2012) Regulation of inflammatory breast cancer cell invasionthrough Akt1/PKBα phosphorylation of RhoC GTPase. Mol Cancer Res10(10): 1306-1318. doi:10.1158/1541-7786.MCR-12-0173. PubMed:22896661. 29. Siegel R, Ward E, Brawley O, Jemal A (2011) Cancer statistics, 2011:the impact of eliminating socioeconomic and racial disparities onpremature cancer deaths. CA Cancer J Clin 4: 212-36. PubMed:21685461. 30. Molina JR, Yang P, Cassivi SD, Schild SE, Adjei AA (2008) Non-smallcell lung cancer: epidemiology, risk factors, treatment, and survivorship.Mayo Clin Proc 83: 584–594. doi:10.4065/83.5.584. PubMed:18452692. 31. Kim Y, Kim E, Wu Q, Guryanova O, Hitomi M et al. (2012) Platelet-derived growth factor receptors differentially inform intertumoral andintratumoral heterogeneity. Genes Dev 26(11): 1247-62. doi:10.1101/gad.193565.112. PubMed: 22661233. 32. Yamashita S, Yamamoto H, Mimori K, Nishida N, Takahashi H et al.(2012) MicroRNAis associated with poor prognosis in colorectalcancer. Oncology 82(4): 205-12. PubMed: 22456107. 33. Wang H, Tan G, Dong L, Cheng L, Li K et al. (2012) CirculatingMiR-125b as a marker predicting chemoresistance in breast cancer.PLOS ONE 7(4): e34210. doi:10.1371/journal.pone.0034210. PubMed:22523546. 34. Garofalo M, Romano G, Di Leva G, Nuovo G, Jeon YJ et al. (2011)EGFR and MET receptor tyrosine kinase-altered microRNA expressioninduces tumorigenesis and gefitinib resistance in lung cancers. NatMed 18(1): 74-82. doi:10.1038/nm.2577. PubMed: 22157681. 35. Schlessinger J (2000) Cell signaling by receptor tyrosine kinases. Cell103: 211-225. doi:10.1016/S0092-8674(00)00114-8. PubMed:11057895. 36. Cully M, You H, Levine AJ, Mak TW (2006) Beyond PTEN mutations:the PI3K pathway as an integrator of multiple inputs duringtumorigenesis. Nat Rev Cancer 6: 184-192. doi:10.1038/nrc1819.PubMed: 16453012. 37. Thomas GV (2006) mTOR and cancer: reason for dancing at thecrossroads? Curr Opin Genet Dev 16: 78-84. doi:10.1016/j.gde.2005.12.003. PubMed: 16359855. 38. Hay N (2005) The Akt-mTOR tango and its relevance to cancer.Cancer Cell 8: 179-183. doi:10.1016/j.ccr.2005.08.008. PubMed:16169463. 39. Samuels Y, Ericson K (2006) Oncogenic PI3K and its role in cancer.Curr Opin Oncol 18: 77-82. doi:10.1097/01.cco.0000198021.99347.b9.PubMed: 16357568. 40. Saxena NK, Sharma D, Ding X, Lin S, Marra F et al. (2007)Concomitant activation of the JAK/STAT, PI3K/AKT, and ERK signalingis involved in leptin-mediated promotion of invasion and migration ofhepatocellular carcinoma cells. Cancer Res 67: 2497-507. doi:10.1158/0008-5472.CAN-06-3075. PubMed: 17363567. 41. Shankar S, Chen Q, Srivastava RK (2008) Inhibition of PI3K/AKT andMEK/ERK pathways act synergistically to enhance antiangiogeniceffects of EGCG through activation of FOXO transcription factor. J MolSignal 3: 7. PubMed: 18355401. 42. Kaller M, Liffers ST, Oeljeklaus S, Kuhlmann K, Röh S et al. (2011)Genome-wide characterization of miR-34a induced changes in proteinand mRNA expression by a combined pulsed SILAC and microarrayanalysis. Mol Cell Proteomics 10(8): M111.010462. PubMed: 43. Silber J, Jacobsen A, Ozawa T, Harinath G, Pedraza A et al. (2012)miR-34a repression in proneural malignant gliomas upregulatesexpression of its target PDGFRA and promotes tumorigenesis. PLOSONE;7: e33844. doi:10.1371/journal.pone.0033844. PubMed:22479456. 44. Nuovo GJ, Nana-Sinkam P, Elton T, Croce C, Volinia S et al. (2009) Amethodology for the combined in situ analyses of the precursor andmature forms of microRNAs and correlation with their putative targets.Nat Protoc; 4: 107-115. doi:10.1038/nprot.2008.215. PubMed:19131963. miR-34a/c Induce TRAIL Sensitivity PLOS ONE | www.plosone.org June | Volume | Issue | eMiR-34a/c-Dependent PDGFR-α/β Downregulation Inhibits Tumorigenesis and Enhances TRAIL-Induced Apoptosis in Lung CancerIntroductionResultsMiR-34a and miR-34c target PDGFR-α and PDGFR-β 3’ UTRsMiR-34a and miR-34c are inversely related to PDGFR-α/β expression in vitro and in vivoMiR-34a and miR-34c overcome TRAIL resistance of NSCLC cells through PDGFR-α and PDGFR-β downregulationPDGFR-α/β downregulation by miR-34a and miR-34c inhibits migration and invasiveness of NSCLC cellsDiscussionMaterials and MethodsLung cancer cell lines and tissue samplesLuciferase assayWestern Blot AnalysisReal-time PCRCell death and cell proliferation quantificationAnti-PDGFR-α and anti-PDGFR-β siRNAs transfectionMiRNA locked nucleic acid in situ hybridization

3 - McDermott-etal.pdf

083937..Ligand-Dependent Platelet-Derived Growth Factor Receptor (PDGFR)-A Activation Sensitizes Rare Lung Cancer and Sarcoma Cells to PDGFR Kinase Inhibitors Ultan McDermott,1Rachel Y. Ames, 1A. John Iafrate, 2Shyamala Maheswaran, 1Hannah Stubbs, Patricia Greninger,1Kaitlin McCutcheon, 1Randy Milano, 1Angela Tam, 1Diana Y. Lee, Laury Lucien,1Brian W. Brannigan, 1Lindsey E. Ulkus, 1Xiao-Jun Ma, 3Mark G. Erlander, Daniel A. Haber,1Sreenath V. Sharma, 1and Jeffrey Settleman 1Center for Molecular Therapeutics, Massachusetts General Hospital Cancer Center and Harvard Medical School, Charlestown,Massachusetts; 2Molecular Diagnostics Laboratory, Department of Pathology, Massachusetts General Hospital andHarvard Medical School, Boston, Massachusetts; and 3AviaraDx, Inc., Carlsbad, California Abstract Platelet-derived growth factor (PDGF) receptors (PDGFR) andtheir ligands play critical roles in several human malignan-cies. Sunitinib is a clinically approved multitargeted tyrosinekinase inhibitor that inhibits vascular endothelial growthfactor receptor, c-KIT, and PDGFR, and has shown clinicalactivity in various solid tumors. Activation of PDGFRsignaling has been described in gastrointestinal stromaltumors (PDGFRA mutations) as well as in chronic myeloidleukemia (BCR-PDGFRA translocation), and sunitinib canyield clinical benefit in both settings. However, the discoveryof PDGFR activating mutations or gene rearrangements inother tumor types could reveal additional patient populationswho might benefit from treatment with anti-PDGFR therapies,such as sunitinib. Using a high-throughput cancer cell linescreening platform, we found that only of tested humantumor-derived cell lines show significant sensitivity to single-agent sunitinib exposure. These two cell lines [a non–small-cell lung cancer (NSCLC) and a rhabdomyosarcoma] showedexpression of highly phosphorylated PDGFRA. In the suniti-nib-sensitive adenosquamous NSCLC cell line, PDGFRAexpression was associated with focal PFGRA gene amplifica-tion, which was similarly detected in a small fraction ofsquamous cell NSCLC primary tumor specimens. Moreover, inthis NSCLC cell line, focal amplification of the gene encodingthe PDGFR ligand PDGFC was also detected, and silencingPDGFRA or PDGFC expression by RNA interference inhibitedproliferation. A similar codependency on PDGFRA and PDGFCwas observed in the sunitinib-sensitive rhabdomyosarcomacell line. These findings suggest that, in addition togastrointestinal stromal tumors, rare tumors that showPDGFC-mediated PDGFRA activation may also be clinicallyresponsive to pharmacologic PDGFRA or PDGFC inhibition.[Cancer Res 2009;69(9):3937–46] Introduction Sunitinib is a multitargeted tyrosine kinase inhibitor thatpotently inhibits vascular endothelial growth factor (VEGF),platelet-derived growth factor (PDGF), and c-KIT receptor kinases(1). In renal cell carcinoma, sunitinib showed superiority overstandard IFN-a therapy (2); sunitinib is now recommended forpreviously untreated patients with advanced renal cell carcinoma.Sunitinib is also approved for treatment of imatinib-refractorygastrointestinal stromal tumors (GIST), many of which harboractivating c-KIT or PDGF receptor (PDGFR) kinase domainmutations (3). A recent phase II clinical study has revealed efficacyof single-agent sunitinib in advanced non–small-cell lung cancer(NSCLC) patients (4). Accumulating evidence indicates thatinhibition of VEGF signaling using various antiangiogenic agentscan suppress tumor growth and improve patient survival (2, 5, 6);however, it is unclear from studies involving multikinase inhibitors,such as sunitinib, as to the relative contribution of VEGF receptorinhibition in suppressing tumor growth.The PDGFR/PDGF system includes two receptors (PDGFRA and PDGFRB) and four ligands (PDGFA, PDGFB, PDGFC, and PDGFD;ref. 7). Ligand binding induces receptor dimerization, enablingautophosphorylation of specific tyrosine residues and subsequentrecruitment of a variety of signal transduction molecules (8).PDGFR regulates normal cellular growth and differentiation (9),and expression of activated PDGFR promotes oncogenic transfor-mation (10), suggesting that activating mutations or generearrangements could play a role in human tumorigenesis.Numerous in vitro and in vivoin vitro and in vivo studies studies showed that inhibition ofPDGFRA signaling disrupts cancer cell survival in the subset ofGISTs with activating PDGFRA mutations (11, 12). In a recent studyof NSCLC patient samples, activated PDGFRA was detected in13% of cases (13), suggesting that a subset of these patients mightbenefit from therapies directed against PDGFRA. Moreover,PDGFRA overexpression has been observed in metastatic versusnonmetastatic medulloblastoma patient samples, and disruptingPDGFRA function inhibited the metastatic potential of medullo-blastoma cells in vitro (14).We recently reported the development of a high-throughput platform for profiling a large panel of human cancer cell lines withmolecularly targeted inhibitors to identify subsets with significantsensitivity (15). That analysis revealed several examples ofgenotype-associated sensitivities to selective kinase inhibitors,showing the utility of this strategy to reveal cell autonomoustumor cell responses to anticancer agents. Here, we describe theprofiling of cancer cell lines for sensitivity to single-agent Note: Supplementary data for this article are available at Cancer Research Online(http://cancerres.aacrjournals.org/). Requests for reprints: Jeffrey Settleman, Center for Molecular Therapeutics,Massachusetts General Hospital Cancer Center and Harvard Medical School, 13thStreet, Charlestown, MA 02129. Phone: 617-724-9556; Fax: 617-726-7808; E-mail:Settleman@helix.mgh.harvard.edu. IAmerican Association for Cancer Research.doi:10.1158/0008-5472.CAN-08www.aacrjournals.org Cancer Res 2009; 69: (9). May 1, Research Article Research. on September 4, 2017. © American Association for Cancercancerres.aacrjournals.org Downloaded from Published OnlineFirst April 14, 2009; DOI: 10.1158/0008-5472.CAN-08 sunitinib, using a monoculture format that precludes anycontribution of drug effects on angiogenesis. Our studies revealedthat the majority of tested cell lines are highly refractory tosunitinib. Of the two cell lines showing sunitinib sensitivity, bothwere found to express high levels of PDGFRA and PDGFC mRNAand phosphorylated PDGFRA protein. ShRNA knockdown ofPDGFRA was as effective as sunitinib in decreasing cellproliferation in both cell lines, and targeting the PDGFC ligandalone was similarly effective.Our findings suggest that whereas antiangiogenesis activity probably accounts for the majority of the majority of the clinical benefit associatedwith sunitinib treatment in solid tumors, in rare cases, beyondPDGFRA-mutant GISTs, activated PDGFRA may be the criticaltarget, and that selective PDGFRA inhibitors may be useful in theclinical management of a subset of tumors that exhibit PDGFRAactivation. Moreover, in tumors with evidence of PDGFC ligandoverexpression, neutralizing antibodies may be an equally effectivetherapeutic modality. Materials and Methods Human cancer cell lines and cell viability assays. Human cancer celllines were obtained from commercial vendors and were maintained and tested for viability using an automated platform, as previously described(15). Cells were treated for h with Amol/L sunitinib and then assayedfor cytostatic or cytotoxic responses. We elected to use this concentration based on steady-state plasma concentrations of fAmol/L at clinicallyrecommended doses of sunitinib in patients and based on the experimentaltime points addressed in the studies. Protein detection. Immunodetection of proteins following SDS-PAGEwas done using standard protocols. Equal protein loading was assessed using a h-tubulin antibody (Sigma). Akt, extracellular signal–regulatedkinase 1/(Erk1/2), phospho-Erk1/(T202/Y204), PDGFRA, phospho- PDGFRA (Y720), phospho-PDGFRA (Y754), signal transducer and activator of transcription (STAT3), and phospho-STAT(S727) antibodies werefrom Cell Signaling Technology. The phospho-Akt (S473) antibody was from BioSource International. All antibodies were used at 1:1,dilution, except h-tubulin (1:10,000).Kinase inhibitors. Sunitinib was obtained from MGH pharmacy. Sorafenib and imatinib were purchased from American Custom Chemicals Corporation. The in vitro kinase specificity profile of all three compounds is listed in Supplementary Table S1. Fluorescence in situ hybridization. Fluorescence in situ hybridization(FISH) was done as described previously (16). Probes for PDGFRA and c-KIT were derived from BAC clones RP11-58C(PDGFRA) and RP11-977G(c-KIT) and purchased from Invitrogen. DNA sequencing. Genomic DNA was isolated using the Gentrapurification system. PDGFRA, PDGFRB , and c-KIT coding sequences were amplified from genomic DNA by PCR. PCR products were purified and subjected to bidirectional sequencing by using BigDye v(AppliedBiosystems) in combination with an ABIsequencer (Applied Bio- systems). Primers used for sequencing are listed in Supplementary Table S2. Electropherograms were analyzed by using Sequence Navigator software (Applied Biosystems). All mutations were confirmed by at least twoindependent PCR amplifications. Cell cycle analysis. Cells were pulsed with Amol/L bromodeoxyur-idine (BrdUrd) for to h before collection, centrifuged, and fixed in ice- cold 70% ethanol. Cells were washed with PBS/0.5% bovine serum albumin(BSA) and incubated in denaturing solution (mol/L HCl) for min at room temperature. After a further wash with PBS/0.5% BSA, the cells were resuspended in mol/L sodium borate for min at room temperature.After an additional wash, cells were suspended with anti-BrdUrd monoclonal antibody for min (1:500; Becton Dickinson). Cells were washed in PBS/0.5% BSA and the pellet was resuspended in FITC- conjugated antimouse IgG (1:50; Vector Laboratories) for min. After an additional wash in PBS/0.5% BSA, cells were stained with Ag/mLpropidium iodide (Sigma) and treated with RNase A (Sigma) before two- dimensional fluorescence-activated cell sorting analysis using CellQuest software (Becton Dickinson). SNP and gene expression analyses. Gene copy numbers weredetermined as previously described as previously described using the GeneChip Human Mapping 250K. The array was then scanned on the GeneChip Scanner 7G and analyzed using GTYPE version with the Dynamic Model Mapping Algorithm and the GeneChip Human Mapping 500K Set library files(Mapping 250K_Nsp). For gene expression studies, RNA was extracted using the Qiagen RNA easy kit (P/N 74106) and amplified and biotin labeled with the Arcturus RiboAmp RNA Amplification Kit using biotinylated ribonucleotides (Perkin-Elmer PN Biotin-11-UTP, NEL543001EA/Biotin-11-CTP, NEL542001EA) during in vitro transcription. Labeled aRNA was hybridized to Affymetric GeneChip Human X3P (GPL1352) using protocols described within theAffymetrix GeneChip Expression Analysis Technical Manual (PNRev. 3). Data were acquired using the Affymetrix GeneChip Scanner with autoloader and 7G upgrade. GCOS ver software was used to run the scanner and analyze the data. The expression value for each gene wascalculated using Affymetrix GeneChip software and data were analyzed using dChip software(17). Probe sets were filtered using two criteria: (a) coefficient of variation between and 1,and (b) P call rate in arrays z20%.Quantitative PCR. Total RNA was isolated and purified from cells using STAT(Tel-Test, Inc.). cDNA was transcribed from Ag of total RNA usingthe AffinityScript Multi Temperature cDNA Synthesis kit (Stratagene).Quantitative PCR was done using the QuantiTect SYBR Green PCR kit (Qiagen) and with an ABI PRISM real-time cycler (Applied Biosystems). Quantification was based on standard curves for each primer set from a serial dilution of the NCI-Hcell line cDNA. All samples were analyzedin triplicate. Primers sequences were GAPDH F, GAGTCAACG- GATTTGGTCGT; GAPDH R, TTGATTTTGGAGGGATCTCG; PDGFRA F, AAATTGTGTCCACCGTGATCT; PDGFRA R, AGGCCAAAGTCACA- GATCTTC; PDGFC F, AACGGAGTACAAGATCCTCAGC; and PDGFC R,CCATCACTGGGTTCCTCAAC. RNA interference studies. ShRNAs targeting sequences within the genesencoding either PDGFRA (n = 10) or its ligand PDGFC (n = 5) wereexpressed from the pLKO.lentiviral vector (Supplementary Table S3). NCI- Hand Acells were infected in the presence of polybrene (Ag/mL).A cell line showing sunitinib-insensitivity (A549) was used to determine infection efficiency based on puromycin resistance and to confirmspecificity. Protein lysates and RNA were collected h postinfection, and cell numbers were determined h postinfection. PDGFC neutralizing antibody experiments. Cells were seeded in1% fetal bovine serum medium and treated the following day with tong/mL of an anti-PDGFC neutralizing antibody (R&D Systems, Inc.). Normal goat IgG at ng/mL concentration was used as a control. Cells were fixed and stained d after treatment, and cell viability was measured as previously described (15). Results Rare human cancer cell lines are sensitive to single-agentsunitinib treatment. Using an automated platform to examinedrug sensitivity in cancer cell lines (15), we tested the sunitinibsensitivity of established human cancer cell lines derived froma wide variety of solid tumor types (Supplementary Fig. S1; ref. 1).Cells were treated for hours with Amol/L sunitinib and thenassayed for cytostatic or cytotoxic responses. Whereas the vastmajority of tested cell lines were largely refractory to treatment,two cell lines (Arhabdomyosarcoma and NCI-HNSCLC) http://biosun1.harvard.edu/complab/dchip/ Cancer Research Cancer Res 2009; 69: (9). May 1, www.aacrjournals.org Research. on September 4, 2017. © American Association for Cancercancerres.aacrjournals.org Downloaded from Published OnlineFirst April 14, 2009; DOI: 10.1158/0008-5472.CAN-08 displayed significant sunitinib sensitivity, as indicated by a >50%reduction in cell number (Fig. 1A). We note that cell lines derivedfrom GISTs, which show clinical sunitinib sensitivity, reflectinginhibition of mutationally activated PDGFR or c-KIT kinases, were absent from the panel of tested lines. A few additional lines showeda relatively weaker response to sunitinib.The sunitinib-sensitive NSCLC-derived cell line harbors focal PDGFRA gene amplification. Among NSCLC cell lines Figure 1. A, pie chart representation of the sensitivity of human cancer cell lines to treatment with Amol/L sunitinib. The drug effect was calculated as thefraction of untreated cells present after h of treatment. The color scheme corresponds to the relative inhibitory effect of treatment, with ratios reflecting the number ofcells remaining following exposure to inhibitor. Details about the most sensitive cell lines identified are shown in the chart, and the cell lines are shown in order ofdecreasing sensitivity (top to bottom). B, pie chart representation of the sensitivity of the NSCLC cell lines to Amol/L sunitinib. Copy number data were generatedfrom 250K Nsp SNP array profiles (or FISH, as indicated by asterisk). Table 1. Elevated PDGFRA copy number in a subset of NSCLC cell lines Chromosome Gene NCI-HNCI-HNCI-HNCI-HNCI-HPDGFRA KIT PDGFC PDGFRB PDGFD PDGFB NOTE: Copy numbers >are in boldface. Data were derived from Affymetrix Nsp 250K SNP array data from NSCLC cell lines. Activated PDGFRA Sensitizes Cancer Cells to Kinase Inhibition www.aacrjournals.org Cancer Res 2009; 69: (9). May 1, Research. on September 4, 2017. © American Association for Cancercancerres.aacrjournals.org Downloaded from Published OnlineFirst April 14, 2009; DOI: 10.1158/0008-5472.CAN-08 tested, significant sunitinib sensitivity was observed only in theadenosquamous NCI-Hline (Fig. 1B). SNP array data availablefor of these lines revealed that NCI-Hcells harbor focalPDGFRA gene amplification (Fig. 1B). This was confirmed byinterphase FISH analysis (Supplementary Fig. S2A). There was noevidence of either c-KIT or PDGFRB genomic amplification orprotein expression in these cells (data not shown). Sequenceanalysis of the entire coding sequence of PDGFRA, PDGFRB , andc-KIT in this cell line revealed a single mutation in exon ofPDGFRA (S478P), within the extracellular domain, which would notbe expected to result in PDGFR activation. The SNP array data revealed similarly elevated PDGFRA genecopy number in four other NSCLC cell lines (NCI-H1693, NCI-H2085, NCI-H23, and NCI-H661); however, these lines weresunitinib insensitive (Table 1; Fig. 2A). FISH analyses of these celllines confirmed PDGFRA amplification (Supplementary Fig. S3).However, analysis of the transcriptional expression profile oftyrosine kinase signaling pathway–associated genes in the 90NSCLC cell line panel revealed that only NCI-Hshowedsignificant expression of PDGFRA mRNA (Supplementary Fig. S4).Furthermore, when the gene expression profile of NCI-Hcellswas compared with the other NSCLC cell lines for the most Figure 2. A, SNP array analysis ofchromosome for the five NSCLC celllines exhibiting elevated PDGFRA copynumber shows increased PDGFC ligandcopy number (5.93) in NCI-Hcells.The blue tracing indicates the degree ofamplification of each SNP in the array.The red line underlying the blue tracingindicates copy number of 2. B, PDGFRAand PDGF ligand mRNA expression inNSCLC cell lines. NCI-Hisindicated by red lettering. Probe sets forligand PDGFB are excluded following afilter based on P call rate in arrays <20%.C, relative PDGFRA mRNA expressionlevels in NSCLC and rhabdomyosarcomacell lines as determined by quantitativereverse transcription-PCR. Cancer Research Cancer Res 2009; 69: (9). May 1, www.aacrjournals.org Research. on September 4, 2017. © American Association for Cancercancerres.aacrjournals.org Downloaded from Published OnlineFirst April 14, 2009; DOI: 10.1158/0008-5472.CAN-08 significant up-regulated and down-regulated mRNA transcripts, themost highly expressed mRNA in NCI-Hcells corresponded toPDGFRA ( fold change of 213; Table 2). When we focused on thoseprobes involved in PDGFR signaling, none of the other four NSCLCcell lines with increased PDGFRA copy number displayed increasedPDGFRA mRNA expression (Fig. 2B). The observed increase inPDGFRA mRNA expression in the NCI-Hcells was confirmedby quantitative PCR (Fig. 2C).Sunitinib dose-response curves for the NCI-Hcell line versus a panel of NSCLC cell lines with normal (Fig. 3A) or increasedPDGFRA gene copy number (Fig. 3B) confirmed the uniquesensitivity in NCI-Hcells. Moreover, expression of phosphor-ylated and total PDGFRA protein was only detected in NCI-H1703cells (Fig. 3A and B), and PDGFRA protein was not detected in anyof the sunitinib-insensitive cell lines. In fact, when we extended thispanel to include an additional NSCLC sunitinib-insensitive celllines, we were unable to detect expression of PDGFRA in any otherlines (Supplementary Fig. S5). Therefore, the increased transcrip-tional expression of PDGFRA in NCI-Hresults in increasedPDGFRA protein and is associated with elevated phospho-PDGFRA, which potentially mediates sensitivity to sunitinib.To assess PDGFRA amplification in clinical NSCLC cases, we analyzed NSCLC primary tumor specimens by FISH anddetected of (3.7%) cases of focal PDGFRA amplification insquamous cell carcinomas (Supplementary Fig. S2B). PDGFRAamplification was not detected in any of adenocarcinoma casesanalyzed. Thus, focal PDGFRA gene amplification arises atrelatively low frequency in NSCLC and may be more common inthe squamous cell setting. Inhibition of PDGFRA activation in NCI-Hcells disruptsdownstream signaling. Treatment of NCI-Hcells withsunitinib for hours resulted in complete inhibition of PDGFRAprotein phosphorylation as well as that of Akt, a PDGFR effector(Fig. 3C). Sunitinib had no effect on such signaling in the sunitinib-insensitive cell lines (data not shown). To verify that PDGFRA-dependent signaling was indeed the basis for the observedsunitinib sensitivity of NCI-Hcells, we treated the cells withtwo additional PDGFRA kinase inhibitors, sorafenib and imatinib.Both compounds exhibited a similar activity to that of sunitinib(Fig. 3C), whereas none of the sunitinib-insensitive NSCLC celllines displayed sensitivity to either agent (data not shown).Furthermore, like sunitinib, both compounds suppressed Aktsignaling in NCI-Hcells (Fig. 3C). Together, these findingssuggest that the NCI-HNSCLC cells are dependent onactivated PDGFRA signaling.To investigate the underlying mechanism for the ability of sunitinib to reduce cell number in NCI-Hcells, we examinedPARP cleavage, an indicator of apoptosis, and cell cycle profile.There was no evidence of PARP cleavage following treatment withAmol/L sunitinib at 24, 48, or hours in this cell line (data notshown), whereas cell cycle analysis confirmed a significant S-phasearrest at each of these time points (Supplementary Fig. S6),consistent with a cytostatic response to drug exposure.PDGFRA activation is associated with sensitivity to sunitinib in a rhabdomyosarcoma cell line. As described above, in theinitial screen of cell lines for sunitinib sensitivity, arhabdomyosarcoma cell line, A-204, was the most highly drug-sensitive line detected (Fig. 1A). To determine whether the Table 2. The most highly up-regulated and down-regulated mRNAs in the NCI-Hcell line compared with all of the NSCLCcell lines Gene Chromosome Fold change LBFC UBFC PDGFRA 4qPDGFRA 4qFLT: fms-related tyrosine kinase 5q10.52FGFR: fibroblast growth factor receptor 8pSHC: SHC transforming protein 1qPLCE: phospholipase C, epsilon 10qSEMA3C 7q1.94HMGA: high mobility group AT-hook 6pHMGA: high mobility group AT-hook 6pEGFR : epidermal growth factor receptor 7p���1.91EGFR : epidermal growth factor receptor 7p���1.88VEGF : vascular endothelial growth factor 6p���2.14MET : met proto-oncogene 7q���2.94DDR: discoidin domain receptor family, member 6p���3.07RGS: regulator of G-protein signaling 2, kDa 1q���4.42MET : met proto-oncogene 7q���4.24EGFR : epidermal growth factor receptor 7p���4.53IRS: insulin receptor substrate 2q���4.78EPS: epidermal growth factor receptor pathway substrate 12q���6.15IRS: insulin receptor substrate 2q���NOTE: Gene expression data were available for of the NSCLC cell lines screened with sunitinib. Genes were included if the fold change was >or<1.2. All data were analyzed using the dChip software. Abbreviations: LBFC, the lower bound of the 90% confidence intervals of fold change; UBFC, the upper bound of the 90% confidence intervals of fold change. Activated PDGFRA Sensitizes Cancer Cells to Kinase Inhibition www.aacrjournals.org Cancer Res 2009; 69: (9). May 1, Research. on September 4, 2017. © American Association for Cancercancerres.aacrjournals.org Downloaded from Published OnlineFirst April 14, 2009; DOI: 10.1158/0008-5472.CAN-08 observed sensitivity could be extended to other rhabdomyosarco-ma cell lines, a panel of six additional rhabdomyosarcoma lineswere tested for sunitinib sensitivity (Fig. 4A). Of the tested lines,only Ashowed sunitinib sensitivity, and in only this cell linewas PDGFRA protein detectable (Fig. 4A, lane 1). Unlike in NCI-Hcells, FISH analysis did not reveal PDGFRA gene amplifica-tion in any of these lines (Fig. 4B), and DNA sequence analysis ofPDGFRA, PDGFRB , and c-KIT in Acells did not reveal any mutations. However, as in NCI-Hcells, we detected asubstantial (15-fold) increase in PDGFRA mRNA expression in Acells (Fig. 2C). Moreover, sunitinib treatment completelyabolished Akt signaling in this line compared with a sunitinib-insensitive rhabdomyosarcoma line (Fig. 4C). In addition, treat-ment of Acells with sorafenib and imatinib also disrupted Aktsignaling (Fig. 4D) and similarly inhibited proliferation (data notshown). These results suggest that rare rhabdomyosarcoma cells Figure 3. Dose-response curves showing the effect of sunitinib on cell numbers h after treatment for NCI-Hand a panel of NSCLC cell lines with eithernormal (A ) or increased (B ) PDGFRA copy number and immunoblots corresponding to these same cell lines showing total PDGFRA and phospho-PDGFRA levels.C, dose-response curves showing the effect of the additional PDGFR inhibitors imatinib and sorafenib on cell numbers h after treatment in the NCI-Hcellline. Immunoblots showing the effect of treating the NCI-Hcell line for h with the indicated concentrations of sunitinib, imatinib, and sorafenib on phosphorylationof PDGFRA and the downstream effectors STATand Akt. Note that p-STATlevels are largely unaffected by drug treatment, whereas p-Akt levels are reduced. Cancer Research Cancer Res 2009; 69: (9). May 1, www.aacrjournals.org Research. on September 4, 2017. © American Association for Cancercancerres.aacrjournals.org Downloaded from Published OnlineFirst April 14, 2009; DOI: 10.1158/0008-5472.CAN-08 are dependent on activated PDGFRA signaling, associated withincreased expression of PDGFRA mRNA.Amplification of the gene encoding the PDGFRA ligand PDGFC mediates PDGFRA activation. Gene expression profilesof NSCLC cell lines using a filtered list of genes involved inPDGFR signaling revealed that NCI-Hwas the only linedisplaying significant transcriptional up-regulation of PDGFRAtogether with the gene encoding one of its ligands, PDGFC(Fig. 2B). The increased PDGFC mRNA in NCI-Hcells (and inAcells) was confirmed by quantitative PCR (SupplementaryFig. S7A). Only the NSCLC cell line NCI-H(sunitinib insensitive)showed similarly elevated PDGFC mRNA, but in the absence ofexpression of PDGFRA mRNA or protein. Further analysis of SNParray data from NSCLC lines revealed a unique coamplificationof the PDGFRA (4q12) and PDGFC (4q32) genes on chromosome 4in NCI-Hcells, which was not observed in any of the other celllines (Fig. 2A).ShRNA-mediated knockdown of PDGFRA and PDGFC was used to directly assess their functional requirement in both the NCI-Hand Acell lines (Supplementary Fig. S7B). There was noeffect of these shRNAs on a sunitinib-insensitive cell line (A549)that lacks PDGFRA expression. In contrast, knockdown of PDGFRAin NCI-Hand Acells significantly reduced proliferation to a similar extent as sunitinib treatment (Fig. 5A). Furthermore,knockdown of PDGFC expression also reduced proliferation inboth lines, and the observed decrease was of the same magnitudeseen following sunitinib treatment (Fig. 5A).We also examined the activity of a neutralizing anti-PDGFC antibody to confirm the ligand knockdown findings and to assessthe potential therapeutic value of anti-PDGFC antibodies in suchtumor cells. We treated three cell lines (A549, NCI-H1703, andA-204) with a concentration range of the anti-PDGFC antibody.Whereas there was no detectable effect on the proliferation of A549cells, the antibody reduced proliferation in the NCI-HandAcell lines to a similar extent to that seen following sunitinibtreatment (Fig. 5B). Notably, the effect was observed in Acellseven at the lowest antibody concentration, potentially reflectingrelatively higher PDGFC expression in the NCI-Hcells(Supplementary Fig. S7A). Combining sunitinib and the anti-PDGFC antibody did not result in any additive inhibitory effects onthese cells (data not shown). ShRNA-mediated depletion ofPDGFRA and PDGFC was used to determine the effect on PDGFRAactivation and downstream signaling in the NCI-Hcells(Fig. 5C). ShRNA-mediated depletion of both receptor and ligandresulted in decreased PDGFRA phosphorylation and inhibition ofAkt and Erk1/phosphorylation. Together, these results indicate Figure 4. A, dose-response curves showing the effect of sunitinib on cell numbers h posttreatment for several rhabdomyosarcoma cell lines. Right, immunoblotsshowing expression of phospho-PDGFRA and total PDGFRA in these lines, with h-tubulin as a loading control. B, FISH analysis of sunitinib-sensitive Acellsusing PDGFRA (RP11-58C6; red) and c-KIT probes (RP11-977G3; green ). C, immunoblots showing the effect of treating the A(sunitinib-sensitive) and A673(sunitinib-resistant) cell lines for h with Amol/L sunitinib on phosphorylation of PDGFRA and the downstream effectors STATand Akt. D, effect of treating Aforh with the PDGFR inhibitors sunitinib, imatinib, and sorafenib on phosphorylation of PDGFRA and the downstream effectors STAT3, Akt, and Erk1/2. Activated PDGFRA Sensitizes Cancer Cells to Kinase Inhibition www.aacrjournals.org Cancer Res 2009; 69: (9). May 1, Research. on September 4, 2017. © American Association for Cancercancerres.aacrjournals.org Downloaded from Published OnlineFirst April 14, 2009; DOI: 10.1158/0008-5472.CAN-08 that both of the sunitinib-sensitive cell lines show a similardependency on increased PDGFRA and PDGFC expression forsustained proliferation. Discussion Our cancer cell line profiling analysis with the multikinaseinhibitor sunitinib has revealed that drug sensitivity in amonoculture context is restricted to a small number of linesexhibiting activated PDGFRA signaling. Moreover, in these cells,PDGFRA activation is coupled to critical downstream effectorssuch as Akt, and disrupting these pathways seems to mediate theinhibitory effects of sunitinib on proliferation. Previous reports ofPDGFRA activation in cancer have been largely confined to GISTs(activating PDGFRA mutations) and rare cases of idiopathichypereosinophilic syndrome (FIP1L1-PDGFRA fusion transcripts;refs. 18, 19). Our findings suggest that in additional rare cases ofNSCLC and sarcoma, PDGFRA activation may be important inmaintaining the malignant phenotype.The clinical success of sunitinib in renal cancer has been suggested to reflect its role as a VEGF receptor inhibitor and theconsequent effects on angiogenesis. Notably, renal cancers arehighly vascularized tumors, suggesting a potential critical require- ment for angiogenesis in that disease setting. However, the abilityof sunitinib to target additional kinases, such as PDGFR, mightcontribute to its clinical activity in renal cancer. We note that ourcell line panel included renal cancer cell lines, none of whichshowed significant sunitinib sensitivity. This suggests that PDGFRis not likely to provide a critical dependency signal in renal cancer;however, a contributing role of PDGFR inhibition in the clinicalactivity of sunitinib cannot be excluded. Whereas in a conventionalxenograft model, any observed consequence of drug treatmenton tumor growth could potentially reflect direct effects on tumorcells as well as effects on the stroma and vasculature, ourmonoculture-based platform provides a means to isolate the tumorcell–autonomous drug response.In both of the sunitinib-sensitive cancer cell lines identified, PDGFRA activation seems to be mediated by increased expressionof the receptor as well as one of its ligands, PDGFC. This is incontrast to other models of receptor tyrosine kinase activationassociated with gene amplification, wherein ligand-independentactivation is typically postulated (20, 21). In NCI-Hcells,activation of the PDGFRA signaling pathways is a consequence offocal PDGFRA and PDGFC gene coamplification. To our knowledge,this is the first report in NSCLCs of overexpression of both anoncogenic receptor tyrosine kinase and its ligand, although Figure 5. A, lentiviral-delivered shRNAs were used to target GFP (negative control), PDGFRA , and PDGFC in A549, NCI-H1703, and Acells, and cellnumbers were measured h postinfection. Treatment with Amol/L sunitinib served as positive control. B, A549, NCI-H1703, and Acells were treated with aneutralizing anti-PDGFC antibody (2.5–ng/mL) or normal IgG control antibody (at ng/mL). Cell numbers were measured d posttreatment. C, effect ofshRNA-mediated depletion of PDGFRA and PDGFC in NCI-Hcells on PDGFRA phosphorylation and downstream signaling h (PDGFRA) and h (PDGFC)postinfection. Cancer Research Cancer Res 2009; 69: (9). May 1, www.aacrjournals.org Research. on September 4, 2017. © American Association for Cancercancerres.aacrjournals.org Downloaded from Published OnlineFirst April 14, 2009; DOI: 10.1158/0008-5472.CAN-08 coexpression of the PDGFRA or PDGFRB receptors with theircognate ligands at higher levels than seen in adjacent normal tissuehas been reported for some gliomas and osteosarcomas (22, 23).Intriguingly, targeting PDGFC had a greater effect on proliferationin both sunitinib-sensitive cell lines than targeting PDGFRA. Thisraises the possibility that PDGFRA is not the sole critical target ofPDGFC in these cells.Our findings also suggest that antibodies directed against PDGFR ligands may have therapeutic potential in PDGFRA-dependent cancer. Traditionally, therapeutic antibodies have beentargeted to cell surface receptors implicated in tumor cellproliferation or maintenance, rather than against their cognateligands (24). Such antibodies typically show a more favorabletoxicity profile than small-molecule kinase inhibitors, and whenconsidered in the context of significant toxicities associated withsunitinib, our findings suggest potential clinical advantages ofantibody-mediated targeting of the PDGFC ligand in some cancers.Our observation that the PDGFRA gene amplification in the NCI- Hadenosquamous NSCLC cell line was also seen in a subset ofsquamous cell NSCLC clinical samples but in none of theadenocarcinoma samples screened by FISH raises the possibilitythat this represents an oncogenic mechanism unique to thishistologic subtype. In agreement with our findings, Rikova andcolleagues (13) detected PDGFRA activation using a phospho-proteomic screen in eight NSCLC patient samples as well as in theNCI-Hcell line. Whereas NSCLC adenocarcinoma patients arebeing actively recruited into clinical trials of epidermal growthfactor receptor (EGFR) tyrosine kinase inhibitors (in the setting ofactivating EGFR mutations) and anaplastic lymphoma kinaseinhibitors (ALK translocations), to date, no drug-sensitizinggenotypes have been identified for squamous cell NSCLC patients(25, 26). It remains to be seen whether retrospective analyses ofsunitinib-responsive NSCLC patients will reveal enrichment forPDGFRA gene amplification or expression, and whether suchpatients’ tumors show squamous histology.Curiously, PDGFRA expression was only detected in the NCI- HNSCLC cells, despite the fact that four other cell linesshowed increased PDGFRA gene copy number. Thus, focalamplification of the PDGFRA gene may uniquely yield high levelPDGFR expression, potentially reflecting an additional genomicalteration within this locus that influences the regulatory regions of PDGFRA gene transcription. Similarly, it remains unclear as to themolecular mechanism underlying increased PDGFRA or PDGFCmRNA expression in the Acells. Sarcomas often harborchromosomal translocations giving rise to oncogenic activation,and these can affect PDGFR signaling. For example, dermatofi-brosarcoma protuberans and giant cell fibroblastomas harborchromosomal rearrangements involving chromosome and 22, inwhich the collagen type Ia(COLIA1) gene undergoes fusion withthe gene encoding PDGFB (27). In one study of cases of uterinesarcoma, 70% of tumors displayed increased PDGFRA expressioncompared with that seen in adjacent normal tissue (28). Likewise,in a study of osteosarcoma patients, increased PDGFRA andPDGFRB expression was observed in tumors in more than 75%of cases (29). Notably, most Ewing sarcomas are associated witha gene fusion that produces a transcription factor (EWS/FLI-1)that promotes PDGFC mRNA expression (30). However, imatinibtherapy in this setting shows minimal clinical activity (31). In thesetumors, which are notoriously refractory to chemotherapy,targeting PDGFR signaling pathways may provide a usefulalternative therapy.In summary, our findings show that ligand-mediated activation of PDGFRA signaling may be a critical mediator of cell proliferationin a small subset of NSCLCs and rhabdomyosarcomas and maysensitize these cancer cells to either selective small-moleculePDGFR kinase inhibitors or ligand-neutralizing antibodies. Ourfindings suggest that sunitinib as well as other PDGFR kinaseinhibitors may provide genotype-associated clinical benefit beyondthe setting of PDGFR-mutant or c-KIT-mutant GISTs. Disclosure of Potential Conflicts of Interest No potential conflicts of interest were disclosed. Acknowledgments Received 11/17/08; revised 1/27/09; accepted 2/25/09; published OnlineFirst 4/14/09.Grant support: National Cancer Institute Specialized Program of Research Excellence in Lung Cancer award PCA090578-06.The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordancewith U.S.C. Section solely to indicate this fact. We thank the members of the Settleman laboratory for helpful discussionthroughout the course of these studies, and Michelle Longworth for assistance withthe cell cycle analysis. Activated PDGFRA Sensitizes Cancer Cells to Kinase Inhibition www.aacrjournals.org Cancer Res 2009; 69: (9). May 1, References 1. Mendel DB, Laird AD, Xin X, et al. In vivo antitumoractivity of SU11248, a novel tyrosine kinase inhibitortargeting vascular endothelial growth factor and plate-let-derived growth factor receptors: determination of apharmacokinetic/pharmacodynamic relationship. ClinCancer Res 2003;9:327–37. 2. Motzer RJ, Hutson TE, Tomczak P, et al. Sunitinibversus interferon a in metastatic renal-cell carcinoma.N Engl J Med 2007;356:115–24. 3. Demetri GD, van Oosterom AT, Garrett CR, et al.Efficacy and safety of sunitinib in patients withadvanced gastrointestinal stromal tumour after failureof imatinib: a randomised controlled trial. Lancet 2006;368:1329–38. 4. Socinski MA, Novello S, Brahmer JR, et al. Multicenter,phase II trial of sunitinib in previously treated, advancednon–small-cell lung cancer. J Clin Oncol 2008;26:650–6. 5. Hurwitz H, Fehrenbacher L, Novotny W, et al.Bevacizumab plus irinotecan, fluorouracil, and leuco-vorin for metastatic colorectal cancer. N Engl J Med2004;350:2335–42. 6. Sandler A, Gray R, Perry MC, et al. Paclitaxel-carboplatin alone or with bevacizumab for non-small-cell lung cancer. N Engl J Med 2006;355:2542–50. 7. Andrae J, Gallini R, Betsholtz C. Role of platelet-derived growth factors in physiology and medicine.Genes Dev 2008;22:1276–312. 8. Valius M, Kazlauskas A. Phospholipase C-gandphosphatidylinositol kinase are the downstreammediators of the PDGF receptor’s mitogenic signal. Cell1993;73:321–34. 9. Ross R, Raines EW, Bowen-Pope DF. The biology ofplatelet-derived growth factor. Cell 1986;46:155–69. 10. Gazit A, Igarashi H, Chiu IM, et al. Expression of thenormal human sis/PDGFcoding sequence inducescellular transformation. Cell 1984;39:89–97. 11. Heinrich MC, Corless CL, Demetri GD, et al. Kinasemutations and imatinib response in patients withmetastatic gastrointestinal stromal tumor. J Clin Oncol2003;21:4342–9. 12. Corless CL, Schroeder A, Griffith D, et al. PDGFRAmutations in gastrointestinal stromal tumors: frequency,spectrum and in vitro sensitivity to imatinib. J ClinOncol 2005;23:5357–64. 13. Rikova K, Guo A, Zeng Q, et al. Global survey ofphosphotyrosine signaling identifies oncogenic kinasesin lung cancer. Cell 2007;131:1190–203. 14. MacDonald TJ, Brown KM, LaFleur B, et al.Expression profiling of medulloblastoma: PDGFRA andthe RAS/MAPK pathway as therapeutic targets formetastatic disease. Nat Genet 2001;29:143–52. 15. McDermott U, Sharma SV, Dowell L, et al.Identification of genotype-correlated sensitivity toselective kinase inhibitors by using high-throughputtumor cell line profiling. Proc Natl Acad Sci U S A 2007;104:19936–41. 16. McDermott U, Iafrate AJ, Gray NS, et al. Genomicalterations of anaplastic lymphoma kinase may sensitizetumors to anaplastic lymphoma kinase inhibitors.Cancer Res 2008;68:3389–95. 17. Lin M, Wei LJ, Sellers WR, Lieberfarb M, Wong WH,Li C. dChipSNP: significance curve and clustering ofSNP-array-based loss-of-heterozygosity data. Bioinfor-matics 2004;20:1233–40. 18. Heinrich MC, Corless CL, Duensing A, et al. PDGFRAactivating mutations in gastrointestinal stromal tumors.Science 2003;299:708–10. Research. on September 4, 2017. © American Association for Cancercancerres.aacrjournals.org Downloaded from Published OnlineFirst April 14, 2009; DOI: 10.1158/0008-5472.CAN-08 Cancer Research Cancer Res 2009; 69: (9). May 1, www.aacrjournals.org 19. Cools J, DeAngelo DJ, Gotlib J, et al. A tyrosine kinasecreated by fusion of the PDGFRA and FIP1Lgenes as atherapeutic target of imatinib in idiopathic hyper-eosinophilic syndrome. N Engl J Med 2003;348:1201–14. 20. Smolen GA, Sordella R, Muir B, et al. Amplifica-tion of MET may identify a subset of cancers withextreme sensitivity to the selective tyrosine kinaseinhibitor PHA-665752. Proc Natl Acad Sci U S A2006;103:2316–21. 21. Brennan PJ, Kumagai T, Berezov A, Murali R, GreeneMI. HER2/neu: mechanisms of dimerization/oligomer-ization. Oncogene 2000;19:6093–101. 22. Sulzbacher I, Traxler M, Mosberger I, Lang S, Chott A.Platelet-derived growth factor-AA and -a receptorexpression suggests an autocrine and/or paracrine loopin osteosarcoma. Mod Pathol 2000;13:632–7. 23. Guha A, Dashner K, Black PM, Wagner JA, Stiles CD. Expression of PDGF and PDGF receptors in humanastrocytoma operation specimens supports the exis-tence of an autocrine loop. Int J Cancer 1995;60:168–73. 24. Adams GP, Weiner LM. Monoclonal antibody therapyof cancer. Nat Biotechnol 2005;23:1147–57. 25. Sequist LV, Martins RG, Spigel D, et al. First-linegefitinib in patients with advanced non-small-cell lungcancer harboring somatic EGFR mutations. J Clin Oncol2008;26:2442–9. 26. Soda M, Choi YL, Enomoto M, et al. Identification ofthe transforming EML4-ALK fusion gene in non-small-cell lung cancer. Nature 2007;448:561–6. 27. Patel KU, Szabo SS, Hernandez VS, et al. Dermatofi-brosarcoma protuberans COL1A1-PDGFB fusion isidentified in virtually all dermatofibrosarcoma protu-berans cases when investigated by newly developedmultiplex reverse transcription polymerase chain reac- tion and fluorescence in situ hybridization assays. HumPathol 2008;39:184–93. 28. Adams SF, Hickson JA, Hutto JY, Montag AG,Lengyel E, Yamada SD. PDGFR-a as a potentialtherapeutic target in uterine sarcomas. Gynecol Oncol2007;104:524–8. 29. Kubo T, Piperdi S, Rosenblum J, et al. Platelet-derivedgrowth factor receptor as a prognostic marker and atherapeutic target for imatinib mesylate therapy inosteosarcoma. Cancer 2008;112:2119–29. 30. Zwerner JP, May WA. PDGF-C is an EWS/FLI inducedtransforming growth factor in Ewing family tumors.Oncogene 2001;20:626–33. 31. Bond M, Bernstein ML, Pappo A, et al. A phase IIstudy of imatinib mesylate in children with refractory orrelapsed solid tumors: a Children’s Oncology Groupstudy. Pediatr Blood Cancer 2008;50:254–8. Research. on September 4, 2017. © American Association for Cancercancerres.aacrjournals.org Downloaded from Published OnlineFirst April 14, 2009; DOI: 10.1158/0008-5472.CAN-08 2009;69:3937-3946. Published OnlineFirst April 14, 2009.Cancer Res Ultan McDermott, Rachel Y. Ames, A. John Iafrate, et al. Sarcoma Cells to PDGFR Kinase Inhibitors Activation Sensitizes Rare Lung Cancer andα(PDGFR)-Ligand-Dependent Platelet-Derived Growth Factor Receptor